Background: Advanced stages of primary melanomas and metastatic lesions are characterized by independence of cells in vitro from the synergistic growth factors required by normal pigment cells. Our studies have shown that a) normal melanocytes in vitro depend on the presence of multiple selected growth factors for survival, proliferation, migration and differentiation; b) these synergistic growth factors not only initiate but also augment and prolong the activated state of several signal transduction intermediates in normal melanocytes; c) two types of signaling proteins are expressed constitutively in melanomas but not in normal pigment cells: one is bFGF (basic Fibroblast Growth Factor), which is also the most abundant melanocyte mitogen in human skin; the other is a group of constitutively tyrosyl-phosphorylated proteins, including pp125 and pp95 that bind Src homology 2 domains (SH2) of Src, and the 85 kDa subunit (p85) of phosphatidyl-inositol-3-kinases (P13K). The melanoma phosphoproteins may or may not be identical to those activated in normal melanocytes by synergistic growth factors. Hypothesis: The stagewise progression to malignancy in pigment cells is due in part to the inappropriate production of bFGF and expression of constitutively active intermediates in signal transduction enabling the cell to short-circuit the need for exogenous synergistic ligands. The aims directed to test this hypothesis are: Specific Aim #1. To pinpoint in vitro the acquisition of growth-factor independence in melanocytes from progressive stages of early melanoma lesions, and to confirm dependency or the loss thereof in vivo by the use of reagents that neutralize the critical ligand(s) from the environment. Specific Aim #2. To identify melanoma pp125 and pp95 by the use of antibodies to known signal transduction proteins and by cloning. Three methods of cloning will be used: 1) sequential affinity-column purifications through GST- SH2, MET-docking peptide domain and anti-phosphotyrosyl antibody, followed by partial amino-acid sequencing and screening of cDNA expression libraries with oligonucleotides; 2) screening of cDNA expression libraries with SH2 probes; and 3) techniques that rely on protein-protein interaction in vivo in bacterial and eukaryotic cells. Clones will be sequenced and the extent of similarity to known genes determined by data base comparison. Specific Aim #3. To characterize the function of candidate melanoma gene products and to determine the importance of these proteins to the transformed phenotype. The enzymatic functions of the cloned phosphoproteins will be investigated, associations with other signaling molecules identified, and expression in normal and malignant cells determined. To assess whether these proteins modify proliferation in vitro and tumorigenicity in vivo, the cloned cDNAs will be inserted into mammalian expression vectors and transfected into normal melanocytes. Significance: The proposed experiments will permit identification of the mission synergist to bFGF in melanomas and may lead to the detection of one or several novel signaling proteins.